Menu
×
John L. Dart Library
Closed for Maintenance
Phone: (843) 722-7550
West Ashley Library
9 a.m. - 5 p.m.
Phone: (843) 766-6635
Folly Beach Library
9 a.m. - 2 p.m.
*open the 2nd and 4th Saturday
*open the 2nd and 4th Saturday
Phone: (843) 588-2001
Edgar Allan Poe/Sullivan's Island Library
Closed for renovations
Phone: (843) 883-3914
Wando Mount Pleasant Library
9 a.m. - 5 p.m.
Phone: (843) 805-6888
Village Library
9 a.m. - 1 p.m.
Phone: (843) 884-9741
St. Paul's/Hollywood Library
9 a.m. - 5 p.m.
Phone: (843) 889-3300
Otranto Road Library
9 a.m. - 5 p.m.
Phone: (843) 572-4094
Mt. Pleasant Library
9 a.m. – 5 p.m.
Phone: (843) 849-6161
McClellanville Library
9 a.m. – 1 p.m.
Phone: (843) 887-3699
Keith Summey North Charleston Library
9 a.m. - 5 p.m.
Phone: (843) 744-2489
John's Island Library
9 a.m. - 5 p.m.
Phone: (843) 559-1945
Hurd/St. Andrews Library
9 a.m. - 5 p.m.
Phone: (843) 766-2546
Miss Jane's Building (Edisto Library Temporary Location)
9 a.m. – 1 p.m.
Phone: (843) 869-2355
Dorchester Road Library
9 a.m. - 5 p.m.
Phone: (843) 552-6466
Baxter-Patrick James Island
9 a.m. - 5 p.m.
Phone: (843) 795-6679
Main Library
9 a.m. - 5 p.m.
Phone: (843) 805-6930
Bees Ferry West Ashley Library
9 a.m. - 5 p.m.
Phone: (843) 805-6892
Mobile Library
Closed
Phone: (843) 805-6909
Today's Hours
John L. Dart Library
Closed for Maintenance
Phone: (843) 722-7550
West Ashley Library
9 a.m. - 5 p.m.
Phone: (843) 766-6635
Folly Beach Library
9 a.m. - 2 p.m.
*open the 2nd and 4th Saturday
*open the 2nd and 4th Saturday
Phone: (843) 588-2001
Edgar Allan Poe/Sullivan's Island Library
Closed for renovations
Phone: (843) 883-3914
Wando Mount Pleasant Library
9 a.m. - 5 p.m.
Phone: (843) 805-6888
Village Library
9 a.m. - 1 p.m.
Phone: (843) 884-9741
St. Paul's/Hollywood Library
9 a.m. - 5 p.m.
Phone: (843) 889-3300
Otranto Road Library
9 a.m. - 5 p.m.
Phone: (843) 572-4094
Mt. Pleasant Library
9 a.m. – 5 p.m.
Phone: (843) 849-6161
McClellanville Library
9 a.m. – 1 p.m.
Phone: (843) 887-3699
Keith Summey North Charleston Library
9 a.m. - 5 p.m.
Phone: (843) 744-2489
John's Island Library
9 a.m. - 5 p.m.
Phone: (843) 559-1945
Hurd/St. Andrews Library
9 a.m. - 5 p.m.
Phone: (843) 766-2546
Miss Jane's Building (Edisto Library Temporary Location)
9 a.m. – 1 p.m.
Phone: (843) 869-2355
Dorchester Road Library
9 a.m. - 5 p.m.
Phone: (843) 552-6466
Baxter-Patrick James Island
9 a.m. - 5 p.m.
Phone: (843) 795-6679
Main Library
9 a.m. - 5 p.m.
Phone: (843) 805-6930
Bees Ferry West Ashley Library
9 a.m. - 5 p.m.
Phone: (843) 805-6892
Mobile Library
Closed
Phone: (843) 805-6909
Patron Login
menu
Item request has been placed!
×
Item request cannot be made.
×
Processing Request
Learning From Multiple Datasets With Heterogeneous and Partial Labels for Universal Lesion Detection in CT.
Item request has been placed!
×
Item request cannot be made.
×
Processing Request
- Author(s): Yan, Ke1 (AUTHOR) ; Cai, Jinzheng1 (AUTHOR) ; Zheng, Youjing2 (AUTHOR) ; Harrison, Adam P.1 (AUTHOR) ; Jin, Dakai1 (AUTHOR) ; Tang, Youbao1 (AUTHOR) ; Tang, Yuxing1 (AUTHOR) ; Huang, Lingyun3 (AUTHOR) ; Xiao, Jing3 (AUTHOR) ; Lu, Le1 (AUTHOR)
- Source:
IEEE Transactions on Medical Imaging. Oct2021, Vol. 40 Issue 10, p2759-2770. 12p.- Subject Terms:
- Source:
- Additional Information
- Abstract: Large-scale datasets with high-quality labels are desired for training accurate deep learning models. However, due to the annotation cost, datasets in medical imaging are often either partially-labeled or small. For example, DeepLesion is such a large-scale CT image dataset with lesions of various types, but it also has many unlabeled lesions (missing annotations). When training a lesion detector on a partially-labeled dataset, the missing annotations will generate incorrect negative signals and degrade the performance. Besides DeepLesion, there are several small single-type datasets, such as LUNA for lung nodules and LiTS for liver tumors. These datasets have heterogeneous label scopes, i.e., different lesion types are labeled in different datasets with other types ignored. In this work, we aim to develop a universal lesion detection algorithm to detect a variety of lesions. The problem of heterogeneous and partial labels is tackled. First, we build a simple yet effective lesion detection framework named Lesion ENSemble (LENS). LENS can efficiently learn from multiple heterogeneous lesion datasets in a multi-task fashion and leverage their synergy by proposal fusion. Next, we propose strategies to mine missing annotations from partially-labeled datasets by exploiting clinical prior knowledge and cross-dataset knowledge transfer. Finally, we train our framework on four public lesion datasets and evaluate it on 800 manually-labeled sub-volumes in DeepLesion. Our method brings a relative improvement of 49% compared to the current state-of-the-art approach in the metric of average sensitivity. We have publicly released our manual 3D annotations of DeepLesion online. 1 https://github.com/viggin/DeepLesion_manual_test_set [ABSTRACT FROM AUTHOR]
- Abstract: Copyright of IEEE Transactions on Medical Imaging is the property of IEEE and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Abstract:
Contact CCPL
Copyright 2022 Charleston County Public Library Powered By EBSCO Stacks 3.3.0 [350.3] | Staff Login
No Comments.